Knowledge Engineering Techniques Research Articles

A knowledge-based architecture for intelligent design support

The development of a knowledge-based design support system is a lengthy and costly process because various computational techniques necessary for intelligent design support are not readily available in a knowledge-based environment. The systematisation of design knowledge needs combined efforts from designers and knowledge engineers. Existing knowledge-based system development tools offer limited support to intelligent design support which require sophisticated knowledge engineering techniques in terms of knowledge representation, inference, control, truth maintenance and learning. In this paper, a knowledge-based architecture for intelligent design support is described. The existing knowledge-based design system architectures are reviewed first. Five key issues in intelligent design support using knowledge engineering techniques, i.e. design knowledge representation, structure of design knowledge base, intelligent control of design process, consistency and context management of design knowledge, and modelling of design collaboration are then discussed. These discussions provide a basis for a description of a knowledge-based design support system architecture which has been implemented in a Lisp-based environment and tested in two different domains. Current application of this architecture in the development of a design support system in the domain of mechanical engineering design at the Cambridge Engineering Design Centre is presented and evaluated.

Special Section Guest Editor's Introduction: Research Surveys On Building Systems With Knowledge And Data Engineering Techniques

Special Section Guest Editor's Introduction: Research Surveys On Building Systems With Knowledge And Data Engineering Techniques

Intelligent musical instrument systems

This paper explores issues related to the application of computer knowledge engineering techniques to musical information, particularly composition. Familiarity with these areas will aid a knowledge engineer in constructing a system for manipulating musical concepts. Knowledge base engineering is examined within the framework provided by Mathews in which his description of an intelligent system for music exploration is considered. Mathews refers to his system as an Intelligent Instrument System . The criteria for an intelligent instrument system is interdisciplinary, and suggest the applicability of artificial intelligence research. This framework is discussed in the context of: a general schema for a musical knowledge base; a brief survey of the field; a discussion, which addresses the feasibility of such systems.

Computer-aided operability study

Operability study is a systematic technique for identifying hazards or operability problems throughout an entire facility. In this paper, we have proposed an approach using knowledge engineering techniques to the automated operability study. The computer-aided operability study system consists of the plant-specific knowledge-base, the generic knowledge-base and the inference engine. Causal relationships between input and output variable deviations for components are modeled using decision tables. Decision tables for components are developed by the user and stored in a generic knowledge-base in computers. The plant structure (piping and instrumentation diagram) and reaction types are inputted to the plant-specific knowledge-base in computers. Each process variable of equipment is examined in sequence by searching the generic knowledge-base, and operability study is generated resulting from the search. We demonstrate via the Solvey process how the computer-aided operability study can identify hazards, and substantiate the usefulness of the method.

Computer-Aided Operability Study for Continuous Processes

Operability Study is a systematic technique for identifying hazards or operability problems throughout an entire facility. In this paper, we have proposed an approach using knowledge engineering techniques to the automated Operability Study. The computer-aided Operability Study System consists of the plant specific knowledge base, the generic data base and the inference engine. Causal relationships between input variable deviations and output variable deviations for components are modeled using decision tables. Decision tables for components are developed by the user and stored in generic data base in computers. The plant structure (Piping and Instrumentation diagram) and reaction types are inputted to the plant specific knowledge base in computers. Each process variable of equipment is examined in sequence by searching the generic data base, and Operability Study are generated resulting from the search.We demonstrate via solvey process how computer-aided Operability Study can identify hazards, and substantiate usefulness of the method.

An expert system for identifying steels and cast irons

This paper presents an application of Knowledge Engineering techniques to the problem of identifying a steel or cast iron from a microphotograph. The essential aim of the implemented system is to help metallography students in the task of learning the concepts relevant for identifying and classifying steels and cast irons. The system has been developed and implemented by means of Knowledge Engineering tools, and all the goals set up at the beginning of the project have been reached.

Computerized systems for on-line management of failures: a state-of-the-art discussion of alarm systems and diagnostic systems applied in the nuclear industry

Abstract It is now well perceived in the nuclear industry that improving plant information systems is vital for enhancing the operational safety of nuclear power plants. Considerable work is underway worldwide to support operators' decision-making, particularly in their difficult tasks of managing process anomalies on-line. The work includes development of (1) advanced alarm systems, such as various kinds of computer-based alarm processing systems, Critical Function Monitoring System, Success Path Monitoring System and Safety Assessment System II, and (2) real-timer diagnostic systems, such as Disturbance Analysis System, Maryland Operator Advisory System II, Model-Integrated Diagnostic Analysis System, Diagnosis System using Knowledge Engineering Technique, Detailed Diagnosis, and Operator Advisor System. This paper presents a state-of-the-art review of plant information systems for on-line management of failures in nuclear power plants, focusing on the methodological features of computerized alarm systems and diagnostic systems.

Engineering of a decision support system for underground coal mine planning and production scheduling

Abstract This paper reports on the development of a computer-based management information system at Datong Coal Company, the largest subsidiary of China National Coal Corporation. Particularly presented is the development methodology of a decision support system for underground coal mine planning and production scheduling, including an integration of computer graphics, database management, mathematical modelling and knowledge engineering techniques and a coupling of life cycle and prototyping software engineering processes to support systems quality assurance.

Engineering experts critics for cooperative systems

AbstractKnowledge collection systems often assume they are cooperating with an unbiased expert. They have few functions for checking and fixing the realism of the expertise transferred to the knowledge base, plan, document or other product of the interaction. The same problem arises when human knowledge engineers interview experts. The knowledge engineer may suffer from the same biases as the domain expert. Such biases remain in the knowledge base and cause difficulties for years to come.To prevent such difficulties, this paper introduces the reader to “critic engineering”, a methodology that is useful when it is necessary to doubt, trap and repair expert judgment during a knowledge collection process. With the use of this method, the human expert and knowledge-based critic form a cooperative system. Neither agent alone can complete the task as well as the two together.The methodology suggested here offers a number of extensions to traditional knowledge engineering techniques. Traditional knowledge engineering often answers the questions delineated in generic task (GT) theory, yet GT theory fails to provide four additional sets of questions that one must answer to engineer a knowledge base, plan, design or diagnosis when the expert is prone to error. This extended methodology is called “critic engineering”.

A knowledge‐based system for envisioning network flow

AbstractThis paper proposes a system called NEVIS (network envisioning system), which envisions a network flow by using a knowledge engineering technique based on qualitative reasoning. NEVIS models the network based on queueing theory. For the server that is the major component of the network, its state is defined for each domain knowledge, and the transition between states is defined based on a change of flow by generation of an event. The state transition is estimated by qualitative reasoning. To avoid ambiguity, quantitative information is added so that an accurate result of reasoning is obtained. The function of NEVIS are to estimate the propagation of the effect of the event by introducing a time concept, to generate a display and explanation for the change of state, and to improve the state of the network through an interaction with its user. Those functions help the user's understanding about the operation of the network. The effective use of the network as well as the means to cope with the event can be planned reflecting the request from the user.

A knowledge-based decision support tool to increase the effectiveness of purchased material quality verifications

In response to the worldwide proliferation of fraudulent materials, a prototype knowledge-based decision support tool has been developed to increase the effectiveness of incoming purchased material quality verifications. Traditional quality verification practices, especially those used in government regulated industries, are ineffective against poor quality materials that appear to be authentic. Safety considerations abound as these highly defective materials are used in systems that are critical to public safety. Quality assurance programs in several instances have failed to stop the influx of fraudulent materials. In many cases only physical inspections and tests can guard aginst defective purchased materials. Unfortunately, it is impractical and cost ineffective for most organizations to inspect all purchased materials. This paper describes how knowledge engineering techniques can be applied to assist in solving this problem. By formulating commonly available and diverse industry information, a knowledge based system becomes a tool that allows users to focus quality verification efforts on highly suspect materials.

Effects of Using a Diagnostic Rule-Based Expert System Developed for the Nuclear Industry

This experiment, conducted at the OECD Halden Reactor Project, Halden, Norway in the spring 1991, aimed to assess the effect on nuclear power plant operators diagnostic behaviour when using a rule based diagnostic expert system. The rule based expert system used in the experiment is called DISKET (Diagnosis System Using Knowledge Engineering Technique) and was originally developed by the Japan Atomic Energy Research Institute (JAERI). The experiment was performed in the Halden man-machine laboratory using a full scope pressurized water reactor simulator. Operator performance in terms of quality of diagnosis is improved by the use of DISKET. The use of the DISKET system also influences operators problem solving behaviour. The main difference between the two experimental conditions can be characterised as while the DISKET users during the diagnosis process are following a strategy which is direct and narrowed, the non-DISKET users are using a much broader and less focused search when trying to diagnose a disturbance.

EFD: A hybrid knowledge/statistical-based system for the detection of fraud

EFD (Electronic Fraud Detection) assists Investigative Consultants in the Managed Care & Employee Benefits Security Unit of the Travelers Insurance Companies in the detection and preinvestigative analysis of healthcare provider fraud. the task EFD performs, scanning a large population of health insurance claims in search of likely fraud, has never been done manually. Furthermore, the available database has few positive examples. Thus, neither existing knowledge engineering techniques nor statistical methods are sufficient for designing the identification process. to overcome these problems, EFD uses knowledge discovery techniques on two levels. First, EFD integrates expert knowledge with statistical information assessment to identify cases of unusual provider behavior. the heart of EFD is 27 behavioral heuristics, knowledge-based ways of viewing and measuring provider behavior. Rules operate on them to identify providers whose behavior merits a closer look by the Investigative Consultants. Second, machine learning is used to develop new rules and improve the identification process. Pilot operations involved analysis of nearly 22 000 providers in six metropolitan areas. the pilot is implemented in SAS Institute's SAS® System, AICorp's Knowledge Base Management System (KBMS®), and Borland International's Turbo Prolog®. © 1992 John Wiley & Sons, Inc.

Knowledge engineering approaches in developing expert simulation systems

Two approaches to knowledge engineering are demonstrated using the soybean insect pest management expert simulation system SMARTSOY. SMARTSOY incorporates the soybean crop growth simulation model SOYGRO as part of its knowledge base. The positivistic approach to knowledge engineering attempts to duplicate thought processes of domain experts in order to arrive at conclusions while the normative approach attempts to duplicate the conclusions of domain experts without consideration of their thought processes. These contrasting approaches require different knowledge engineering techniques and validation procedures. Validation, evaluation and calibration of the two approaches are demonstrated using SMARTSOY for 84 field scenarios.

INTEGRATING USER INTERFACE DEVELOPMENT AND MODERN SOFTWARE DEVELOPMENT

A long-term goal for software engineers is integrating the separate processes of user interface development and modern software development. With emergent CASE technology, software engineers can begin to explore ways to achieve this integration. Exploration involves investigating candidate methodologies that let developers apply different development strategies to different parts of an interactive system. Disciplined long-term investigation requires that the fundamental principles governing each process be fixed and that evolving development methods comprising each process be accommodated. This paper proposes a computer-based process model that fixes the principles and accommodates evolving methods. Model features include a collection of software engineering and knowledge engineering techniques that supports a development organization of human and computer-based agents, a coordination activity that supports opportunistic behavior of developers, a unifying representation that leads to mutually consistent results from developers, and an extendable topology that enhances collaboration among developers while reducing their communications burden.

Welborix, a Novel Approach Towards User-Interface Technology in Well-Performance Simulation

Shell Research has developed an integrated well-performance simulation tool that is widely used throughout the Shell Group. However, users do not fully appreciate its advanced functionality because of the cumbersome way in which relevant input data have to be composed. To alleviate this hindrance, the WELBORIX project was initiated, the aim of which was to build an adaptive human-computer interface for an existing wellbore simulator. The main aim of the research demonstrator WELBORIX, however, was to identify general requirements for a system designed to simplify a complex interaction between human and computer. When an experienced user of the simulator models a problem, he or she draws on an extensive body of knowledge. Applying knowledge-engineering techniques, the WELBORIX interface helps a less experienced user to work with the simulator by dealing with the complexity of the input. The compositon of a valid set of input parameters is described in a knowledge base, and the process of the human-computer interaction is controlled by an inference engine that is based on backward chaining. Moreover, facts derived during the computer's reasoning process determine the composition of the output and the sequence in which it is presented to the user. A special mechanism has been designed to obtain any needed factual information that cannot be derived from rules; it consists of a transparent layer connecting the missing facts with user-interface components. The adaptative human-computer interface of WELBORIX enables occasional users to interact with the wellbore simulator in a natural way, calling objects by their common name. The experienced user, on the other hand, does not have to sacrifice any of the system's flexibility; he or she is free to take short cuts. The user controls a session through a direct-manipulation, graphic user interface, which offers combined text and graphics, context-dependent help and multiple representations of data. Object-orientation, an excellent paradigm to support the flexibility such a system demands, was embraced during the design of the interface. Because a generic approach was followed, production-technology systems should benefit from the general-purpose user-interface components that were developed to represent objects relevant to the application domain. For efficiency reasons, WELBORIX was originally developed on a SUN/UNIX workstation using COMMON LISP (with object-oriented extensions) and X-Windows. However, because it has demonstrated the excellent potential offered by knowledge-engineering techniques in combination with object orientation in the area of human-computer interfaces, it is currently being made operational using C++ and OSF/Motif.

Automated selection of mathematical software

Current approaches to recommending mathematical software are qualitative and categorical. These approaches are unsatisfactory when the problem to be solved has features that can “trade-off” in the recommendation process. A quantitative system is proposed that permits tradeoffs and can be built and modified incrementally. This quantitative approach extends other knowledge-engineering techniques in its knowledge representation and aggregation facilities. The system is demonstrated on the domain of ordinary differential equation initial value problems. The results are significantly superior to an existing qualitative (decision tree) system.

RXPERT: a prototype expert system for formulary decision making.

RXPERT, a prototype, computer-based, expert system that models the decision-making processes for an ambulatory (non-hospital) formulary, is described as an example of how expert systems may be used to support pharmacy decision making. Basic information about expert-system technology is provided through this example. Computer-assisted decision making is becoming an important and accepted aspect of complex, health-related decisions. Because expert-system support may become an integral component of future, complex, pharmacy decision making, it is important for pharmacists to become familiar with this technology and its possibilities for supporting pharmacy decisions. Expert systems offer the potential advantages of making the human decision-making process explicit, more consistent, easily duplicated in many locations simultaneously, and easy to update and document. Although an expert system is seldom intended to replace human decision makers, it can provide valuable support for complex, multivariable decisions. Typical knowledge-acquisition and knowledge-engineering techniques, as well as the characteristics and structure of expert systems, are described, relative to the development of the RXPERT prototype. Although RXPERT is not yet in use, the process for using an expert system to support an individual committee member's personal assessment of a drug product is described. Decision-support expert systems are potentially useful to pharmacists in complex decision-making tasks.

知識工学による原子力発電プラントの異常兆候監視支援方式

An architecture to support nuclear power plant operators for monitoring abnormal symptoms has been proposed based on the techniques of knowledge engineering, and the feasibility of a plant monitoring support system was investigated. The purpose of the support system is to present the operators with useful information so that they can make correct judgment at an early and subtle stage of abnormal plant conditions. In the architecture proposed, abductive reasoning is performed to search for causal events and deductive one to predict consequential events using the knowledge representing plant components as frames and those representing causal relations as production rules. A method to deal with uncertainties in each type of reasoning has been adopted, and it is used to rank several hypotheses of causal events and to assess the importance of plant parameters for monitoring. A prototype system was developed, and its usefulness was tested using a case of failure in a recirculation pump of a BWR plant.

Knowledge engineering approach to risk management and decision-making problems

Abstract Application of knowledge engineering techniques is regarded as one of the promising approaches to resolution of inter-export conflicts on risk-related decision problems. Among various causes of the conflicts, differences in unstated assumptions and in subjective weighting of risk-relevant factors are treated as the major issues. To identify the causes and to evaluate their influences, a modified version of analytic hierarchy process (AHP) is proposed in this paper. The effectiveness of the proposed methodology in the conflict resolution is demonstrated through several examples.